Authors:
Paul Elzinga
1
;
Jonas Poelmans
2
;
Stijn Viaene
3
and
Guido Dedene
2
Affiliations:
1
Police Amsterdam-Amstelland, Netherlands
;
2
Faculty of Business and Economics, K.U. Leuven, Belgium
;
3
Vlerick Leuven Gent Management School, Belgium
Keyword(s):
Formal Concept Analysis (FCA), Emergent SOM, Domestic violence, Knowledge discovery in databases, Text mining, Exploratory data analysis.
Related
Ontology
Subjects/Areas/Topics:
Applications of Expert Systems
;
Artificial Intelligence
;
Artificial Intelligence and Decision Support Systems
;
Biomedical Engineering
;
Business Analytics
;
Data Engineering
;
Data Mining
;
Databases and Information Systems Integration
;
Datamining
;
Enterprise Information Systems
;
Health Information Systems
;
Informatics in Control, Automation and Robotics
;
Information Systems Analysis and Specification
;
Intelligent Control Systems and Optimization
;
Knowledge Engineering
;
Knowledge Engineering and Ontology Development
;
Knowledge Management
;
Knowledge-Based Systems
;
Knowledge-Based Systems Applications
;
Ontologies and the Semantic Web
;
Sensor Networks
;
Signal Processing
;
Society, e-Business and e-Government
;
Soft Computing
;
Software Engineering
;
Symbolic Systems
;
Web Information Systems and Technologies
Abstract:
Over 90% of the case data from police inquiries is stored as unstructured text in police databases. We use the combination of Formal Concept Analysis and Emergent Self Organizing Maps for exploring a dataset of unstructured police reports out of the Amsterdam-Amstelland police region in the Netherlands. In this paper, we specifically aim at making the reader familiar with how we used these two tools for browsing the dataset and how we discovered useful patterns for labelling cases as domestic or as non-domestic violence.